210 Sargant. — Formation of the Sexual Nuclei 
referred to another series, cut from absolute alcohol-material 
and stained with Renaut’s haematoxylic eosin (Figs. 24, 28 ). 
The daughter-chromosomes separate exactly as in vegetative 
nuclei. Indeed, this is the most typical of the five homotype 
divisions. The series of diagrams headed e homotype ’ in 
Plate XI are founded on the vegetative karyokinesis (I, p. 451), 
but they might have been drawn equally well from the division 
we are now considering. The longitudinal fission of the 
chromosomes is perfectly clear (Fig. 27). 
Both the resting and the dividing nucleus are invariably 
found close to one wall of the pollen-grain. The spindle is not 
unusually short. Its long axis is at right angles to the cell-wall 
near which it lies (Fig. 28). But during the dispirem-period the 
connecting threads look as if they were crushed between the 
two daughter-nuclei. The inner one is now much nearer than 
before to the outer, which is still close to the cell-wall. 
I have never succeeded in seeing centrosomes, but a dif- 
ferentiated mass of cytoplasm is very conspicuous both at the 
inner pole of the spindle (Fig. 28) and on the inner side of 
the spirem-nucleus (Fig. 24). Radiations can often be traced 
from this mass into the surrounding cytoplasm. The latter 
shows a honeycomb-structure very clearly (Figs. 24, 28), but 
this may possibly be due to imperfect fixing. The exine offers 
considerable resistance to the penetration of fixing agents. 
• The two daughter-nuclei continue to lie very close to each 
other for some time after the connecting threads have dis- 
appeared. A difference in size very soon becomes apparent. 
The generative nucleus is the smaller, and in the mature 
pollen-grain it is enclosed within a lens-shaped naked cell 
(see M. Guignard’s beautiful figure : Nouvelles Ftudes, 
Plate X, Fig. 27). During the interval between the forma- 
tion of two daughter-nuclei and the dehiscence of the anther, 
each pollen-grain swells so that the outline of its equatorial 
section is circular, and it also grows larger (cf. Fig. 30 with 
Fig. 28). No change takes place in the structure of the two 
resting-nuclei until the pollen-grain reaches the stigma where 
it will germinate. 
